The UnAustralian

Sunday, November 09, 2003
Cosmic Rays and Global Warming Redux

A while ago, a discussion came up on Troppo Armadillo (make sure that you read the comments - especially the end where Nir Shaviv jumps in) about work of Nir Shaviv and Jan Veizer who have put forth a theory that the earth's position in galaxy (and hence levels of cosmic rays which it receives) play a significant role on the earth's temperatures.

I wrote a long post on this, which I'm going to reproduce here:

A recent hypothesis put forward by Nir Shaviv and J Veizer (Shaviv 2003) has triggered some interesting posts on Troppo Armadillo. This post is a modified version of a comment I left at the Armadillo.

Their study essentially finds a good correlation between galactic cosmic rays and global temperatures over the last billion years. When the variability of cosmic rays is compared with surface temperatures, the residues (unexplained variability) is low. From this (and also that the correlation between CO2 variability and temperatures on the geological scale is also weak), they suggest that the effect that CO2 has on global warming is also low.

What needs to be remembered is the massive uncertainties involved here. Reconstructing the earth's past, is fraught with error (ironically Michael Mann's reconstruction of the past 1000 years temperatures has been critiqued for it's uncertainties - this reconstruction is six orders of magnitude longer, and the errors tend to grow as we look further back into time).

The study relies upon the following reconstructions being accurate 1) past levels of cosmic rays 2) past temperatures 3) past levels of CO2.

The first reconstruction should be viewed with some skepticism as there is very little data available. As far as I'm aware Shaviv's reconstruction of past cosmic ray flux is the only reconstruction.

The second reconstruction is debatable. Paul Olsen has pointed out that one of the cool periods in the reconstruction is actually considered to be a warm period (Clarke 2003). As Scheffler and co-workers in their study of the past glaciation of the Karoo Basin note:

As the solar system passes through one of the four galactic spiral arms, the cosmic ray flux increases, which may trigger global glaciation... The poor correlation between the cosmic ray flux maximum and the timing of the Pennsylvanian glaciation excludes this external forcing mechanism and underlines the need for further investigation (Scheffler 2003).

The third reconstruction suffers from the same problem. The reconstruction is taken from measurements of isotopes in sedimentary rocks. Gregory Retallack, has suggested that this style of dating can be effected methane to give dodgy results. His reconstruction of past CO2 levels based of stomata, gives a much better fit between temperatures and CO2 (Retallack 2001).

Shaviv has detailed the problems with the first two reconstructions in his paper in New Astronomy (Shaviv 2002).

This isn't to say that the last two reconstructions are wrong - it's just that we don't know. And if any of them is significantly wrong (especially 1 and 2) then the whole hypothesis falls apart (The match between 1 and 2 is quite good, if either of these are wrong, it will significantly increase the residue variability, gives CO2 more room to play a bigger role).

It should be noted that the authors have put forward a case for cosmic rays being a major determent of the earths climate on the geological time scales, which is at least (in my humble uniformed opinion) as strong as any existing theory.

Lets for the sake of argument, assume that the conclusions of the authors are entirely correct. When we look at what this means for the recent 20th century warming, we get a very interesting result.

Contrary to Ken Parish’s statement, that “new research suggesting that much of the observed 20th century global warming is actually caused by variations in the activity of supernovae (rather than carbon dioxide generated by evil western capitalists)" the solar systems transition through the galaxy isn't the cause of the 20th century warming - simply because the solar system's movement over the last 100 years has been next to nothing on an astronomical time-scale. Or as John Quiggin notes “All of this is based on the assumption that you can go directly from a 500 million year time scale to one of 50-100 years”

However, this isn't too say that cosmic rays haven't varied over this time period. Since 1953, we've been keeping records of cosmic radiation flux at several locations. In the decade time period, cosmic radiation flux variability is mostly effected by the sun's magnetic field (ie. it oscillates with the solar cycle) - in fact, cosmic rays have been used in the past to argue as too how the small variations in solar intensity can cause a greater than expected effect on the earth's atmosphere.

As we have good records of cosmic radiation and global temperatures, it is possible to check the correlation between the variation in cosmic ray intensity and global temperatures. Needless to say, the correlation isn't good (since the mid-1980's the cosmic ray flux has increased - this should lead to a decrease in global temperatures, if they dominate, temperatures during this period have increased).

My conclusion is that Shaviv and Veizer have produced a strong case for cosmic rays playing a dominant role in the earth's climate on a geological scale, however, it is unlikely that they play a significant role in the 20th Century warming.


T. Clarke. Galactic Dust Cooling the Earth Nature, July 2003 (news article).

G. Retallack. A 300-Million-Year Record of Atmospheric Carbon Dioxide from Fossil Plant Cuticles Nature, May 2001, vol 411, page 286.

K. Scheffler, S. Hoemes and L. Schwark. Global Changes During Carboniferous-Permian Glaciation of Gondwana: Linking Polar and Equatorial Climate Evolution by Geochemical Proxies Geology, July 2003, vol 31, page 605.

N. Shaviv. The Spiral Structure of the Milky Way, Cosmic Rays, and Ice Age Epochs on Earth New Astronomy, 2002.

N. Shaviv and J. Veizer. Celestial Driver of Phanerozoic Climate GSA Today, July 2003.

Now, the hypothesis has come under some pretty heavy criticism. You can read the initial comments here, and the authors response + a response to the response here.

I'm quite pleased because many of the problems which I brought up are apparently true (always good to know), however I do have some sympathy for the authors. A public statement calling your work crap is never nice, and I thought that Shaviv was quite a sincere person when he post on the Troppo Armadillo thread linked to above. It's a bit sad that this sort of stuff is fought out in public, but I also think that it is unavoidable given the highly polarized debate over global warming.

Given what I've read, the only thing that I would change from the reposted comment above (apart from the appalling spelling and grammar) would be that the case for cosmic rays being the predominant cause of temperature variability on a geological scale is considerable poorer than what I thought.

(Thanks to Quark Soup for some of the links).
| 6:34 PM